SkateFormer: Skeletal-Temporal Transformer for Human Action Recognition
Core Concepts
SkateFormerは、骨格-時間トランスフォーマーであり、効率的な行動認識を実現します。
Abstract
Introduction:
Skeleton-based action recognition is widely used in various scenarios.
Transformer-based methods address limitations of GCNs in capturing relations between joints.
Methodology:
SkateFormer partitions joints and frames based on skeletal-temporal relations.
Utilizes Skate-MSA for selective focus on key joints and frames crucial for action recognition.
Experimental Results:
Outperforms recent state-of-the-art methods across multiple modalities.
Achieves new state-of-the-art in interaction recognition.
Data Extraction:
"Our SkateFormer sets a new state-of-the-art for action recognition performance across multiple modalities (4-ensemble condition) and single modalities (joint, bone, joint motion, bone motion), showing notable improvement over the most recent state-of-the-art methods."
Customize Summary
Rewrite with AI
Generate Citations
Translate Source
To Another Language
Generate MindMap
from source content
Visit Source
arxiv.org
SkateFormer
Stats
"Our SkateFormer sets a new state-of-the-art for action recognition performance across multiple modalities (4-ensemble condition) and single modalities (joint, bone, joint motion, bone motion), showing notable improvement over the most recent state-of-the-art methods."